Universal memory runtime for AI agents
SaferSkills independently audited pensyve (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Use Pensyve as Codex's persistent working memory. The skill is intended for explicit $pensyve invocation from Codex, selection through /skills, /pensyve command workflows, text-level @pensyve compatibility requests, and implicit routing when a task clearly depends on prior project memory.
@pensyve as a mention-style memory request.Do not use this skill for trivial commands, simple formatting, one-off shell output, or facts that should not be persisted.
pensyve_recall with a query that includes the task and entity.pensyve_remember for durable facts, decisions, preferences, and constraints.pensyve_observe for session outcomes, root causes, failed approaches, and procedural lessons.[procedural].pensyve_inspect for entity-specific memory listings.pensyve_status for connection or namespace health.pensyve_forget.@pensyve platform dispatch. Current Codex support is $pensyve, /skills, /pensyve, and text-level @pensyve interpretation by the model.~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.